reassign 541307 src:linux-2.6 2.6.30-1
retitle 541307 agp/intel: gpu lockups on 865G
tag 541307 upstream patch
kthxbye

On Thu, Aug 20, 2009 at 11:27:39 -0700, Carl Worth wrote:

> Excerpts from Brice Goglin's message of Thu Aug 20 00:07:26 -0700 2009:
> > 865G support is severely broken unfortunately. It was supposed to
> > improve when disabling render acceleration in driver 2.8.0-2 but it
> > still miserably fails on my machine. I had to downgrade to 2.7.1 without
> > UXA to get something to work.
> 
> Hi Brice,
> 
> I'm glad to see this message from you. Just moments ago I fired up an
> 865 machine planning to disable RenderAccel for 8xx in the upstream
> code. It's interesting to learn that this isn't enough to fix things,
> (since we know that the RenderAccel code is very broken for 8xx).
> 
> Have you had any luck discovering more about what's happening here,
> (for example, doing any bisecting or so?).
> 
> I'm interested in getting the driver to be stable on all hardware in
> current use as quickly as possible, (even if that means just reverting
> to the old code as in 2.7.1 for 8xx for example).
> 
> Anyway, I'll start experimenting with my 865 here, but any input you
> might be able to provide will be appreciated.
> 
I'm reassigning this to the kernel since it turned out to be a bug in
the intel_agp chipset flush routine.
The patch is queued at
http://git.kernel.org/?p=linux/kernel/git/anholt/drm-intel.git;a=commit;h=e517a5e97080bbe52857bd0d7df9b66602d53c4d
Kernel maintainers, any chance to get this into sid soonish?

Thanks,
Julien



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Reply via email to